Code for ''It HAS to be Subjective: Human Annotator Simulation via Zero-shot Density Estimation''. Please read our paper [arXiv] for detailed descriptions of the proposed human annotator simulation (HAS) method.
PyTorch==1.11.1
speechbrain==0.5.14
normflows==1.6
numpy==1.21.0
scikit-learn==1.0.2
statsmodels==0.13.5
Emotion classification on MSP-Podcast
Prepare label: python3 data_preparation/prep_msp-label.py
Prepare training scp: python3 data_preparation/prep_msp-scp.py
Hate speech detection on HateXplain
python3 data_preparation/prep_hx.py
Speech quality assessment on SOMOS
python3 data_preparation/prep_somos.py
python3 Train_S-CNF.py Train_S-CNF.yaml --output_folder='exp'
python3 Train_I-CNF.py Train_I-CNF.yaml --output_folder='exp'
For S-CNF: python3 scoring_S-CNF.py exp/test_outcome-E{PLACEHOLDER}.npy
For I-CNF: python3 scoring_I-CNF.py exp/test_outcome-E{PLACEHOLDER}.npy
If you find our paper and/or code useful for your research, please consider citing our paper:
@article{wu2023has,
title={It HAS to be Subjective: Human Annotator Simulation via Zero-shot Density Estimation},
author={Wu, Wen and Chen, Wenlin and Zhang, Chao and Woodland, Philip C},
journal={arXiv preprint arXiv:2310.00486},
year={2023}
}